BACKGROUND Exercise testing may be of value in identifying a group of children at high risk of subsequently developing respiratory symptoms. As few longitudinal studies have investigated this issue, the bronchial hyperresponsiveness to exercise in asymptomatic children was evaluated as a risk factor for developing asthma related symptoms in young adulthood.
METHODS A community based sample of 1369 schoolchildren, first investigated in 1985 at a mean age of 9.7 years, was followed up after a mean of 10.5 years. Nine hundred and twenty children (67%) were asymptomatic in childhood and 777 (84.9%) of these were re-investigated at follow up. At the first examination a maximum progressive exercise test on a bicycle ergometer was used to induce airway narrowing. The forced expiratory volume in one second (FEV1) after exercise was considered abnormal if the percentage fall in FEV1 was more than 5% of the highest fall in the reference subjects characterised by having no previous history of asthma or asthma related symptoms. The threshold for a positive test was 8.6% of pre-exercise FEV1.
RESULTS One hundred and three subjects (13%) had wheeze within the last year at follow up and, of these, nine (9%) had been hyperresponsive to exercise in 1985. One hundred and seventy subjects (22%) had non-infectious cough within the previous year, 11 of whom (6%) had been hyperresponsive to exercise in 1985. Multiple regression analysis showed that subjects with hyperresponsiveness to exercise had an increased risk of developing wheeze compared with subjects with a normal response to exercise when the fall in FEV1 after exercise was included as a variable (threshold odds ratio (OR) 2.3 (95% CI 1.1 to 5.5)). The trend was not significant when exercise induced bronchospasm was included as a continuous variable (OR 1.02 (95% CI 0.97 to 1.06)).
CONCLUSIONS Asymptomatic children who are hyperresponsive to exercise are at increased risk of developing new symptoms related to wheezing but the predictive value of exercise testing for individuals is low.
- exercise testing
- young adults
- symptom development
Statistics from Altmetric.com
Asymptomatic bronchial responsiveness seems to be a risk factor for the development of asthma1 2 but only a few studies have investigated whether the risk of new symptoms can be categorised according to exercise performance. In a cross sectional study Busquetset al 3 concluded that the prevalence of asthma related symptoms and bronchial responsiveness to exercise had different meanings in a community survey. However, exercise induced bronchospasm can be measured before the onset of asthmatic symptoms4 and this could reflect a state of “latent asthma” in the airways.
The aim of this study was to investigate whether exercise induced bronchospasm in childhood could predict the development of new asthma related symptoms in later childhood and adolescence.
A cohort of 1369 schoolchildren in third grade participated in the initial study. The mean age was 9.7 years (range 8.4–11.4). The details of the random population selection and examination have previously been published.5 Nine hundred and twenty children (67%) were asymptomatic in childhood, 777 of whom (84.9%) were re-investigated at follow up.
Parents gave informed consent for their children to participate in the baseline study and subjects gave informed consent before participating in the follow up study. The study was approved by the local research ethics committee and the Danish Data Surveillance Authority.
The exercise challenge test in childhood involved maximum progressive exercise on a bicycle ergometer with an increase in work load every third minute until exhaustion.6 Nose clips were not used. The heart rate was measured with a Polar Sport tester. The effort was accepted as maximal when the subject reached a heart rate of 185 beats/min. A McDermott bellows spirometer was used to measure forced expiratory volume in one second (FEV1) before exercise and at 5 and 10 minutes after exercise. The European guidelines for spirometric testing were followed.7 Lung function after the exercise test was expressed as the maximal fall in FEV1 as a percentage of the pre-exercise FEV1measurement. The exercise test was considered abnormal if the percentage fall in FEV1 was more than 5% of the highest fall in the reference subjects characterised by having no previous history of asthma or asthma related symptoms.1 The threshold for a positive test was 8.6%.
All subjects who answered positively in 1985 to at least one of the questions on asthma or the presence of asthma related symptoms were excluded from the analysis. Asthma related symptoms were asked with reference to the previous one year period (see ). The question: “Do you have hayfever?” was used to identify subjects with allergic rhinitis in 1985. A history of asthma in the family was noted if at least one parent or one sibling had asthma. Subjects were labelled smokers if their estimated lifetime tobacco consumption exceeded one pack year. The pack year estimate was calculated by multiplying the tobacco consumption in grams per day divided by 20 with the duration of smoking in years.
χ2 tests and t tests were used to compare participants and non-participants. A multiple logistic regression model was applied with symptom status at follow up as the dependent variables. Two different symptom definitions were used as dependent variables—wheeze and non-infectious cough. The independent variables introduced into the model were bronchial hyperresponsiveness to exercise, sex, allergic rhinitis, tobacco smoking, FEV1% predicted, and history of asthma in the family. Two tailed tests were used with a 5% significance level. The analysis was performed using SPSS-PC+ 7.5.1 (SPSS Inc, Chicago, Illinois, USA).
No statistically significant differences were found between participants and non-participants at follow up with respect to sex, age, birth weight, or the data measured in 1985 (prevalence of allergic rhinitis, physical fitness, height, weight, body mass index, FEV1 and FVC). Hyperresponsiveness to exercise in 1985 and the relation to the development of asthma related symptoms at follow up are shown in table 1. One hundred and three subjects (13%) had wheeze during the last year at follow up of whom nine (9%) had been hyperresponsive to exercise. One hundred and seventy subjects (22%) had non-infectious cough within the previous year, 11 of whom (6%) had been hyperresponsive to exercise in 1985.
The results obtained by multiple regression are shown in table 2. If hyperresponsiveness to exercise was included as a continuous variable in the model it was positively associated with both wheeze and cough 10 years later, but without reaching statistical significance (odds ratios (OR) 1.01 (95% CI 0.97 to 1.06) and 1.01 (95% CI 0.97 to 1.05), respectively.
Previously symptom free children who were hyperresponsive to exercise had a greater risk of developing wheeze in early adulthood than asymptomatic children with a normal response to exercise. Exercise induced bronchospasm had only a weak association with cough and the main contributor to the development of cough was tobacco smoking.
Increased airway responsiveness to various stimuli is a feature of asthma9 and may be determined genetically.10 11 Wheeze seems to be more closely associated with bronchial responsiveness than cough.3 12Haby and coworkers showed that exercise induced bronchoconstriction could be measured before the onset of symptoms in childhood and the degree of responsiveness was positively associated with wheeze attacks during the previous year.4 The sensation of respiratory symptoms may have to exceed a certain “threshold” before respiratory discomfort develops.13 This may explain why hyperresponsiveness to exercise can be detected before asthma related symptoms are reported. However, the association is undoubtedly more complex as a large part of the population will experience wheeze without a demonstrably abnormal fall in FEV1 after exercise.14 This may be partly due to the high variability and low agreement between replicate exercise tests resulting from the high intrinsic variability in airway calibre, particularly in asthmatic subjects.15 16
The maximum progressive exercise test on a bicycle ergometer has mainly been used to measure physical fitness by an indirect measure of maximal oxygen uptake (ml/kg/min) and it has been claimed to have some limitations for assessment of exercise induced bronchospasm.17 Physiologically there is no evidence that the mechanisms of exercise induced airway narrowing differ between a running test and a bicycle test8 17 18 but, when the bicycle test is used to provoke airway narrowing, it has been claimed that bicycle testing seems to be a poorer stimulus than running in provoking an attack of asthma. It has also been claimed that the coefficients of variation between repeated tests vary more than in running tests,17 and that the maximal degree of airway narrowing in subjects occurs at 60–85% of their predicted maximum oxygen consumption. Durations longer than this and expenditure of greater power are claimed to be associated with flattening or diminution of the response. Subsequent studies have not confirmed these observations.18-21 Furthermore, the cut off values obtained to indicate the 5% “most asthmatic” of the reference subjects for an abnormal test corresponded with previous findings in other studies,4 22 23 so we think that our bronchoconstriction test can be used to assess the present research question.
In conclusion, we have shown that asymptomatic children who were hyperresponsive to exercise were at higher risk of developing asthma related symptoms as young adults than children without hyperresponsiveness to exercise. This suggests that abnormal airway changes may precede the onset of asthmatic symptoms. However, the exercise test has insufficient predictive value to be useful as a guide to individual prognosis.
The authors are indebted to Niels Hyldebrandt who encouraged them to start the study. The study was supported by the Danish Lung Association, The Danish Medical Research Council, Odense University, The Hørslev Foundation, K A Rohde and his wife’s Foundation, Else and Mogens Wedell-Wedellsborg Foundation, and CEO Jacob Madsen and his wife Olaga’s Foundation.
In 1985 subjects with asthma were identified by the question: “Have you ever had asthma, i.e. periods of wheeze and/or cough?” and asthma like symptoms by the questions, asked with reference to the previous one year period: “Do you have wheeze and/or cough at rest?”, “Do you have wheeze and/or cough when you are exercising?”, “Do you feel shortness of breath at night?”, “Do you feel shortness of breath in the morning?”, “Do you have wheeze and/or cough in foggy weather?” and “Do you have bronchitis, i.e. periodic cough during several days/weeks?”.
At follow up, asthma was identified by an affirmative answer to the question: “Is it your doctor’s opinion that you have asthma?”. Asthma like symptoms were identified by the questions, asked with reference to the previous one year period: “Do you have attacks of breathing trouble with wheezing or whistling?”, “Do you have trouble breathing at night?”, “Do you have trouble breathing at all?” and “Have you had a period with cough lasting three or more days in succession?”. Cough in relation to colds was only considered if lasting two weeks or more (additional questions asked: “Do you only cough in connection with colds?” and “For how long does the cough period usually last?”).24